WHO meetings on international collaborative research on craniofacial anomalies Possibilities

WHO meetings on international collaborative research on craniofacial anomalies 3 Possibilities for improving the treatment of CFA Three interrelated clinical management issues were identified by participants as being priorities for international collaborative research: the identification and dissemination of optimal clinical interventions for the management of craniofacial anomalies (evidence-based care); the identification and dissemination of strategies to optimize the quality of services that deliver care (quality improvement); and the identification and dissemination of strategies to increase the availability of care to all affected citizens of the world (access and availability). 3.1 Evidence-based care Evidence-based care is considered to be “the integration of best research evidence with clinical expertise and patient values”. In respect of therapeutic interventions, the most powerful evidence is derived from systematic reviews that provide a synthesis of relevant randomized controlled trials (Sackett et al., 2000). However, for CFA care providers there are some challenges ahead. Even for the longest established CFA intervention – the management of cleft lip and palate – the scientific basis of the discipline is weak. Virtually no elements of treatment have been subjected to the rigours of contemporary clinical trial design (Roberts et al., 1991) and there is a bewildering diversity in practices. A recent survey of European cleft services revealed that, in 201 teams, 194 different surgical protocols were followed for unilateral clefts alone (Shaw et al., 2001). Table 4 shows the variation in sequence and number of operations in current use to repair a unilateral cleft in Europe. 14 Global strategies to reduce the health-care burden of craniofacial anomalies Table 4: Sequence of operations for the repair of unilateral complete cleft lip and palate First operation Lip closure Lip closure Lip and hard palate closure Lip and soft palate closure Lip, hard and soft palate closure Lip closure Lip and soft palate closure Lip and alveolar closure Soft palate closure Lip adhesion Lip and alveolar closure Lip adhesion Lip adhesion Hard and soft palate closure and alveoloplasty Lip and soft palate closure Lip adhesion Lip closure Total Soft palate closure Hard palate closure and gingivo-alveoloplasty Hard and soft palate closure Lip and hard palate Lip closure Soft palate closure Lip, hard and soft palate closure Lip and hard palate closure Lip closure Hard palate closure and alveolar bone grafting Lip closure Soft palate closure Hard and soft palate closure Gingivo-alveoloplasty Hard palate closure Soft palate closure Soft palate closure Hard palate closure Hard palate closure Hard palate closure and alveolar bone grafting Second operation Hard and soft palate closure Soft palate closure Soft palate closure Hard palate closure Hard palate closure Third operation Fourth operation % 42.8 15.3 10.4 10.0 5.0 3.5 2.5 2.0 2.0 1.5 1.0 1.0 1.0 0.5 0.5 0.5 0.5 100.00 Source: Shaw et al., 2001 15 WHO meetings on international collaborative research on craniofacial anomalies Generally speaking, choices in surgical technique, timing and sequencing, and choices in ancillary procedures such as orthopaedics, orthodontics, and speech therapy are arrived at after disappointment in the results of former practices, rather than on the basis of firm evidence that the new protocol has succeeded elsewhere. As a consequence, the unsubstantiated testimony of enthusiasts for a particular treatment has done much to shape current practices. Typically, enthusiastic claims are made for a new type of therapy; the procedure is widely adopted; a flow of favourable anecdotal reports ensues; little or no positive evidence develops to support the desirability of the procedure; there is a sharp drop in the number of clinical reports, again without evidence to support the change (Spriestersbach et al., 1973). 3.1.1 Sources of bias in CFA research See Box B, facing page. Bo facing pag age. Not surprisingly then, empirical research frequently demonstrates that in studies of health care interventions without randomization, an inflated view of effectiveness results (Kunz and Oxman, 1998). Thus controlled trials of a series of psychiatric medications found them effective only 25% of the time but, in uncontrolled studies of the same medications, 75% were positive. Even more dramatically, none of a series of randomized trials of portacaval shunt surgery found clear evidence of benefit but 75% of uncontrolled studies did. 3.1.2 The hierarchy of evidence for CFA research As non-randomized studies make up the great majority of the current literature in CFA treatment they must be appraised with great caution, being appreciated for the contributions to knowledge they can make and also recognized for their inherent limitations. They conform to the following broad hierarchy (Roberts et al., 1991): nec re ts: Anecdotal case rep o r ts Case reports may signal important new developments in clinical practice, but the evidence they contain for a widespread change in practice remains generally unconvincing in the absence of subsequent rigorous confirmation. series: Case ser ies Reports of a series of cases treated by the same method provide more substantial evidence of the merits of a particular technique or programme of treatment, and provide the professional 16 Global strategies to reduce the health-care burden of craniofacial anomalies BOX B Sources of bias in CFA research The general rules of “health technology assessment” are well established and the quality of treatment comparisons conforms to a widely accepted hierarchy, from anecdotal reports to randomized trials and systematic reviews. This hierarchy relates to the degree of effort made to minimize ever-present sources of research bias that readily lead to false conclusions. The following certainly apply to the literature concerning CFA, and make comparisons between reports unreliable: Susceptibility bias (lack of equivalence between groups of cases): Some patients will inevitably be more susceptible to the treatment applied, because their condition is less severe or because they inherently possess a better prognosis.Thus the apparent effectiveness of any technique, applied to a group of cases that are inherently more amenable to correction, will be inflated if compared to another technique applied to a more challenging group of cases. For example, comparisons of facial growth data may be dubious where there are inherent differences in facial form between communities. Similarly, speech development may be less good in circumstances where the socioeconomic profile of the population served by a particular centre is less favourable, or where the local spoken language calls for different oro-pharyngeal skills. Proficiency bias: In a similar manner, a more skilled surgeon or clinical team can also inflate the apparent effectiveness of a technique. If operator A is 10% better than operator B, and technique X is 5% better than technique Y, a false conclusion will be reached in a comparison of technique Y performed by A, versus technique X performed by B. Follow-up bias: The consumer of journal or conference reports needs some reassurance that the “whole story” has been given and that follow-up has been as rigorous for the cases that went badly as for those that went well. Without knowing about all the cases on whom a particular technique was tried, reliable conclusions cannot be drawn. Exclusion bias: In reporting the effectiveness of an intervention it is often tempting to exclude cases retrospectively, where the expected progress was not achieved.Typical grounds for retrospective exclusion might be lack of compliance on the part of the patient or suspicion that an underlying condition (e.g. an ill-defined “syndrome”) has prevented the intervention from working. Irregular application of the rules of retrospective exclusion clearly can remove any equivalence that comparison groups may have had. Analysis bias: Given the virtual absence of agreed rating schemes for outcome evaluation, reporting in the CFA literature is inevitably inconsistent. And without objectivity in appraisal – as achieved with blinded, independent panels – comparisons must be unsure. Reporting bias: It would appear that clinical researchers, like pharmaceutical companies, are more likely to report positive findings than negative ones. But not only are findings more likely to be reported if they are positive, but they are also more readily accepted for publication by journals, more readily accepted for conferences, more often published in English, and more often cited in later publications (Easterbrook et al., 1991; Dickersin et al., 1992; Dickersin and Min 1993; Egger et al., 1997; Stern and Simes, 1997). 17 WHO meetings on international collaborative research on craniofacial anomalies Differences arising from biases ... may exceed actual differences attributable to procedures community with a general impression of relative efficacy. Rather commonly, however, outcome is measured in the short term and the enthusiasm of the reporters may impair true objectivity. Thus primary bone grafting, first heralded as an important breakthrough in caseseries reports, was later shown by randomized controlled trials to be harmful to facial growth (Rehrmann et al., 1970; Jolleys and Robertson, 1972). On the other hand, case series of secondary bone grafting using cancellous iliac crest grafts revealed persuasive evidence that one aspect of outcome, the patient’s dentition, could be reliably restored beyond levels previously attainable (Boyne and Sands, 1972, 1976; Bergland et al., 1986). The immediacy of these benefits ruled against the need for a randomized trial though potential growth disturbances still deserved consideration (Semb, 1988). Future trials of bone grafting may, however, still be necessary to examine individual aspects of surgical technique or timing, or to test the suitability of alternative graft materials. Case series rarely provide evidence of the superiority of one technique over others where a choice of broadly similar methods exists and in which any improvement may be modest rather than dramatic. This is a major problem in the evaluation of the primary surgical repair of clefts, since this may be achieved with apparently similar success by methods that differ in technique, timing and sequence. Differences arising from the biases listed above are likely to exceed actual differences attributable to the procedures. n-rand mize compariso ando ison No n-rando mize d co mparison studies: Opportunities for nonexperimental comparisons of therapies or programmes of care can arise in several ways: by the coexistence of different therapies at the same centre, by the replacement of one therapy with another, or by collaboration of two or more centres. In such comparisons attempts may be made to reduce bias. mpariso of co-existing ther ison o-exist Comparison of co-existing ther apies: In using retrospective material, such as case notes or clinical databases, checks can be made on the equivalence of the groups, commonly in terms of gender, age or diagnostic subtype. Preferably, cases can be matched pair-wise on these characteristics, or adjustments can be made in the analysis by stratification or the use of multivariate statistical methods. In either case, however, doubt will remain that important prognostic factors have been masked for, if two or more therapies were being used concurrently within a single centre, selective allocation to treatment may have occurred. For example, decisions as to when (at what age) to perform surgery may be influenced by unrecorded aspects of the condition, the availability of personnel, the health of the child or 18 Global strategies to reduce the health-care burden of craniofacial anomalies parental attitudes and characteristics. Should these factors influence outcome, confounding would occur in any study of the effect of age on surgical outcome. Even if it is possible to match or adjust data to remove bias due to gender, age or severity, this gives no guarantee that some other prognostic factor that may affect outcome is not associated with choice of treatment. And of course, a critical factor in surgical outcome is the differing proficiency of different surgeons. mpariso with histo ison contr Compar ison with histor ical controls: These studies may arise as natural experiments by changes in therapy within a treatment centre. Such research is feasible when durable records (radiographs, study casts, speech recordings, photographs, etc.) are obtained in a standardized way for both those subjects treated by an earlier method (the historical controls) and those subjects treated by a subsequent one, allowing simultaneous evaluation. An alternative circumstance in which such studies arise is where data for a group of patients receiving a standard treatment already exists and can be gathered in a similar way when a new treatment is introduced. This design requires only half the number of patients to be gathered prospectively as a randomized clinical trial and is clearly attractive where recruitment of cases is slow. Furthermore, it has been argued that, in circumstances of poor outcome, it may be unethical to withhold new treatment in order to create a control group (Gehan, 1984). There are nevertheless several biases and possibilities for confounding that generally tend to favour the newly-introduced procedure. In practice, changes in technique at a treatment centre often come about as a result of changes in personnel who may have performed differently in respect of the previous method. This leads to bias due to differences in skill of personnel associated with either treatment method. For example, a new method of treatment is often tested by an experienced and innovative surgeon who may be expected to achieve better results than the average surgeon. This clearly introduces the confounding effect of operator proficiency with treatment. Even where there is stability of staff, bias reflecting gradual changes of ability and technique are highly likely and definition or ascertainment of prognosis may change. New methods may also be initially applied with some selectivity to “suitable” cases as experience is gained. Other aspects of clinical management may have been altered with the intention of improving outcome, creating additional possibilities for bias in favour of the innovative procedure. Multivariate methods have been suggested as a way to adjust for these biases, but serial changes in treatment are likely to take place in parallel, resulting in a strong Changes in technique are often a result of changes in personnel with differences in skill 19 WHO meetings on international collaborative research on craniofacial anomalies association between treatment variables (Semb et al., 1991). This is one reason why historical control design is generally unsuited to evaluating primary cleft surgery since other changes in the total programme of care are likely to have occurred during the extensive recruitment period. The bias favouring the innovative procedure is a major cause for concern with historical control studies as they may either fail to resolve a controversy or alternatively create ethical concerns that preclude further, more rigorous, comparisons. Favourable outcomes suggested for a new procedure by historical control studies have been disputed by subsequent randomized controlled trials (Pinsky, 1984; Pollock, 1986). Thus, the danger exists that historical control studies could set in motion an unwarranted cycle of change with no benefit to the patient and consequently delay the process of development. The reduction in recruitment time for a historical control study in which data are gathered prospectively on a new method is also less important when extended follow-up is required of each case. If, for example, the proposed follow-up of a trial of 2 methods of primary surgery is 10 years and the recruitment time of patients sufficient for a randomized trial is 4 years, the total duration would be 14 years. The potential saving of time in a partially prospective, historical control study would only be 2 years (14%). nte -centr compariso ison Inter-centre co mparison : The multi-centred approach offers distinct advantages for cleft or CFA treatment centres, as the generation of adequate samples within specific subtypes treated by contrasting treatment modalities is extremely difficult. Prospectively planned recall of cases at participating centres allows data on outcome to be collected in a standardized way, and rigorous planning and execution across the centres can ensure consecutive case recruitment and consistent evaluation (Shaw et al., 1992a,b). Provided procedures for entry into the study are equivalent in all participating centres, this strategy is extremely valuable in assessing the outcome of surgery, together with other major components of the treatment programme at respective centres. However, for primary cleft surgery it is difficult, if not impossible, to establish the key beneficial or harmful features of a specific treatment due to the invariably complex and arbitrary mix of surgical technique, timing and sequence, ancillary procedures, and surgical personnel (Shaw et al., 1992b). For example, if two centres differ in the use of presurgical orthopaedics and types of primary lip and palate surgery, there is no way to determine which of these procedures might be responsible for any difference in outcome between centres, nor would a null result Bias favouring innovative procedures is a major cause for concern with historical control studies 20 Global strategies to reduce the health-care burden of craniofacial anomalies allow the conclusion that individual aspects of the treatment programme are equivalent. The method is therefore better suited to comparative clinical audit and quality assurance than definitive clinical research. The existence of significant disparities in outcome of the overall treatment process provides a basis for speculating as to the possible cause, and inter-centre studies should, therefore, be highly motivating towards the generation of specific hypotheses for subsequent trials. andomize controlle trials: led Randomized controlled tr ials For the comparison of therapies there is little doubt that the randomized controlled trial is generally the method of choice, scientifically and ethically. Prognostic factors, including clinical proficiency, whether known or unknown to the investigator, tend to be balanced between treatment groups. Since patients are registered prior to treatment and followed up prospectively according to a clearly defined protocol, missing data are less likely as the potential loss to follow-up and late exclusion is reduced. Formalizing the protocol at the outset, as required by an ethical review board or funding agency, increases the likelihood of impartial analysis. The likelihood of reporting the results is also increased but by no means guaranteed. Randomized controlled trials can, of course, also be performed badly. Notably, if the randomization procedure is not strictly applied (i.e. if allocation is not fully concealed from the investigators), bias can enter. Inadequate concealment in clinical trials is associated with higher odds ratios, i.e. an inflated view of effectiveness emerges (Moher et al., 1998), as in the case of non-randomized studies. Trials with insufficient cases may also give misleading results. stematic re iew rand mize tr ando Systematic re v iew o f rando mize d tr ials: Systematic review of all relevant randomized trials is the optimal method for establishing whether scientific findings are consistent and can be generalized across populations, settings and treatment variations, or whether findings vary significantly by particular subsets. Explicit methods used in systematic reviews limit bias and improve reliability and accuracy of conclusions (Chalmers and Altman, 1995). Meta-analysis – the use of statistical methods to summarize the results of independent trials – can provide more precise estimates of the effects of health care than those derived from individual studies. The Cochrane Collaboration is an international organization established to prepare, maintain and promote the accessibility of systematic reviews of the effects of health-care interventions and, as randomized trials in CFA are completed and reported, it will become a primary source of reviews and dissemination (www.cochrane.org). Systematic review of randomized trials is the optimal method of choice 21 WHO meetings on international collaborative research on craniofacial anomalies 3.1.3 Improving the evidence base for CFA Given the relative scarcity of CFA, the dispersion of clinical services and the diversity of therapies, the establishment of a sound evidence base seems unlikely, without the development of a strategic international framework. Early experience with randomized trials in cleft management Almost thirty years ago, Spriestersbach et al., (1973) identified the need for prospective research to resolve central problems of cleft management, but remarkably few randomized trials have been performed in cleft lip and palate surgery despite being the surest means of advancing the discipline in the face of overwhelming uncertainty about the relative efficacy of countless different programmes of care around the world. In a review of 25 years of the Cleft Palate Journal, only 5 controlled clinical trials were identified, with only 1 involving a follow-up of surgery for more than 4 years (Roberts et al., 1991). Robertson and Jolleys conducted two small randomized controlled trials of primary surgery in the 1960s. In the first study a sample was randomized in respect of alveolar bone grafting at the time of primary surgery in infancy (Robertson and Jolleys, 1968). Follow-up revealed a detrimental effect on facial growth in the grafted group (Robertson and Jolleys, 1983). The second study involved 2 groups of 20 cases where 1 group’s anterior palate closure was delayed until 5-years of age. No benefit for dentofacial growth was found in delaying hard palate closure (Robertson and Jolleys, 1974). A follow-up study when the children were 11 years of age reached the same conclusion (Robertson and Jolleys, 1990). In a quasi-randomized trial (patients entered on basis of birthdates), Wary et al. (1979) found a difference in perioperative morbidity following 3 types of palate repair in 47 patients with a variety of cleft types: V-Y pushback, Langenbeck, Langenbeck with superiorly based pharyngeal flap. Speech outcomes were subsequently reported for 52 patients (Holtman et al.,1984). Morbidity was least with the Langenbeck and speech outcomes were the same in all three. Chowdri et al. (1990) compared rotation-advancement and triangular flaps in unilateral cleft lip repair in 108 cases and found no differences in lip and nose appearance. In another quasi-randomized controlled trial (patients alternated rather than randomized) on speech outcome, Marsh et al. (1989) compared palate repair with or without intravelar veloplasty in 51 subjects with a broad range of palatal cleft types. Speech evaluations were made at a twoyear follow-up. No difference in outcome was detected but the procedure, including intravelar veloplasty, required a significantly longer operating time. Remarkably few randomized trials have been performed in cleft lip and palate surgery 22 Global strategies to reduce the health-care burden of craniofacial anomalies Most trials have involved relatively small samples Another randomized controlled trial on speech outcome and maxillary growth in patients with unilateral complete cleft lip and palate operated on at 6 versus 12 months of age was undertaken in Mexico (Ysunza et al., 1998). The study groups consisted of 41 subjects operated on at 12 months of age, and 35 subjects operated on at 6 months. There was no statistically significant difference in velopharyngeal insufficiency, maxillary arch development or soft tissue profile as measured on cephalometric radiographs. However, phonologic development was significantly better in patients operated at six months and none of the patients in this group developed compensatory articulation. The authors concluded that cleft palate repair performed at six months significantly enhances speech outcome and prevents compensatory articulation disorder. The same group compared minimal incision palatopharyngoplasty with and without individualized velopharyngeal surgery for velopharyngial insufficiency in 72 patients with submucous cleft palate, and found no benefit for the more complex procedures (Ysunza et al., 2001). For patients with velopharyngeal insufficiency (VPI), secondary surgery to the pharynx is often recommended. Whitaker et al. (1972) found no difference in outcome in a randomized trial of 35 patients, comparing superiorly- versus inferiorly-based flaps. More recently, pharyngeal flap or sphincter pharyngoplasty were compared in a multi-site randomized controlled trial of 97 patients. Patients were evaluated before surgery, then 3 and 12 months following surgery, by perceptual speech evaluation, video nasopharyngoscopy, nasometry, polysomnographic sleep study, lateral cephalometric radiographs, audiometry and tympanometry. Preliminary analysis has shown both techniques to be equally effective and equally safe (VPI Surgical Trial Group, 2001). A larger replication of this trial is currently under way at the Hospital for Research and Rehabilitation of Craniofacial Anomalies, University of São Paulo, Brazil. Most of the above trials have involved relatively small samples, but two current surgical trials are taking place on a more ambitious scale. A randomized controlled trial to compare velopharyngeal function for speech outcomes in two groups of patients with complete unilateral cleft lip and palate is also being undertaken at the Hospital for Research and Rehabilitation of Craniofacial Anomalies in Brazil (Williams et al., 1998). The two palatoplasty techniques tested are von Langenbeck with intravelar veloplasty and the Furlow procedure. A total of 608 patients are being entered into 1 of 2 age categories; patients having surgery before 1 year of age and patients undergoing surgery at approximately 1½ years of age. This study is designed to determine which of the two surgical procedures is superior in constructing a velum capable of affecting velopharyngeal competency for the development of normal speech. 23 WHO meetings on international collaborative research on craniofacial anomalies The feasability of randomized trials has been demonstrated Since 1986, North European teams have been developing a concerted programme of multidisciplinary inter-centre research in cleft lip and palate. This includes a comparison of surgical outcome in four Scandinavian centres (Friede et al., 1991; Enemark et al., 1993) and six European centres (Shaw et al., 1992a,b; Mars et al., 1992; Asher-McDade et al., 1992; Mølsted et al., 1992, 1993a,b; Morrant and Shaw, 1996; Grunwell et al., 2000). Following these collaborations, the limitations of inter-centre studies became increasingly obvious to these teams, as it became clear that it would be impossible to separate and compare the single elements of the package of care provided in the different centres. This experience provided a compelling stimulus for starting randomized controlled trials in primary surgery of clefts and 10 centres are currently participating in a set of 3 parallel trials where groups of teams are testing their traditional local protocols against a common protocol. At the time of writing, more than half of the proposed sample of 450 infants with unilateral cleft lip and palate has been entered into this “Scandcleft” trial (Semb, 2001). Randomized trials of other interventions have also been completed. These include a trial of artificial bone (Ping et al., 2001), a trial of nasal floor augmentation (Chen. et al., 1999), trials of anaesthesia or analgesia (Bremerich et al., 2001; Prabhu et al., 1999; Ahuja et al., 1994; Nicodemus et al., 1991), a trial of perioperative steroid therapy (Senders et al., 1999), a trial of perioperative antibiotics (Anland et al., 1995), speech therapy following velopharyngeal surgery (Pamplona et al., 1999), inclusion of mother in speech therapy (Pamplona et al., 2001), phonologic versus articulatory speech intervention (Pamplona et al., 1999), the use or nonuse of presurgical orthopaedics (Kuijpers-Jagtman and Prahl, 1996; Kuijpers-Jagtman and Prahl-Andersen, 1997; Konst et al., 2000; Prahl et al., 2001), the use or non-use of arm splints following surgery (Jigjinni et al., 1993), feeding after surgery (Darzi et al., 1996; Lee et al., 1999), feeding methods in infancy (Brine et al., 1994; Shaw et al., 1999), and the use of continuous airway pressure (CPAP) in the treatment of hypernasality (Kuehn et al., in press), and fluoride supplements for dental caries (Lin and Tsai, 2000). Such efforts demonstrate the feasibility of randomized controlled trials in the CFA field and indicate the probable shape of future progress. Thus trials of sufficient power are likely to be mounted either through collaboration between funding agencies, clinical scientists, and large, high volume centres (possibly in the developing world, as in the Brazilian trials above). Alternatively, they may be mounted as multi-centre investigations within collaborative groups with strong geographic or cultural links, as in the Scandcleft trial. Each will have a place. 24 Global strategies to reduce the health-care burden of craniofacial anomalies Challenges in mounting clinical trials Among the challenges in mounting clinical trials concerned with CFA are, firstly, adequate length of follow-up since interventions are often applied at an early stage of life and their full consequences only revealed some years later; secondly, the location of CFA may impair many structures and functions calling for the quantification and weighting of diverse outcomes. Above all, however, is the challenge of sample size since the various subgroups of CFA occur infrequently. Current estimates suggest that 2 groups of around 75 cases of the same diagnostic subtype are required in trials of cleft surgery. For example, more than 1 million births would have to occur for a trial including 150 infants with complete, nonsyndromic, unilateral complete cleft lip and palate (assuming a rate of 1 per 7 of all cleft types, 1 cleft per 700 births, 75% compliance with all inclusion/exclusion criteria, and consent obtained in 90% of cases). On the basis of the actual rate of entry to the Scandcleft trial mentioned above, smaller countries, such as Denmark (population 5.3 million) and Norway (population 4.4 million) would take 8 and 11 years respectively to recruit 150 cases in a single-nation trial, despite a rate of 1 cleft per 500 births. History indicates that not all surgical innovations are an enduring success Ethical issues in randomized trials The ethical issues raised in randomized trials in CFA care are interesting (Berkowitz, 1995; Shaw, 1995), in particular the double standards that are applied in clinical experimentation. History indicates that not all surgical innovations are an enduring success. Discredited, though once fashionable techniques, include gastric freezing for bleeding peptic ulcer, carotid body denervation for bronchial asthma, portacaval shunt to prevent oesophageal variceal bleeding, nephropexy for viceroptosis, removal of chronically inflamed appendix and periarterial sympathectomy (Baum, 1981; Salzman, 1985). Indeed, numerous reports show that new treatments are as likely to be worse, as they are to be better, than existing alternatives (Chalmers, 1997). Where the doctor leads, however, most patients and parents will follow, raising an important ethical dilemma. If a surgical team wishes to test an innovative procedure in a randomized trial it must obtain ethical approval from an appropriate authority and fully inform each new patient of any uncertainty and/or risk prior to obtaining his/her signed consent. Ironically, if the team wishes to try out the same innovation on all its patients, no such rules currently apply (Chalmers and Lindley, 2000). “Ethical codes that seek to protect patients ... regulate the responsible investigator but not the irresponsible adventurer” (Lantos, 1994). In the United States the National Commission for the Protection of Human 25 WHO meetings on international collaborative research on craniofacial anomalies Subjects recommended that “medical committees should be responsible for ensuring that major innovations undergo proper scientific evaluation” and be charged with “determining which new treatments need to be evaluated, the proper method of evaluation and how to limit the use … prior to the completion of that evaluation” (Tonelli et al., 1996). As yet no such body exists, neither in the United States nor elsewhere. In the light of the above, there exists a strong imperative to mount clinical trials across a range of CFA where true uncertainty of effectiveness (equipoise) exists, and to apply the customary rules for informed consent and ethical approval from appropriate authorities. When trials in a developing country are planned and funded by a developed country, it would offer reassurance if a cooperative or parallel trial were also to be undertaken in the developed country unless, of course, the trial has relevance only for developing countries. There exists a strong imperative to mount clinical trials across a range of CFA Planning for surgical trials Bo facing pag age. See Box C, facing page. Measuring outcome The ultimate goal of CFA care is restoration of the patient, as far as possible, to a “normal” life, unhindered by handicap or disability. However, the measurement of normalcy is a highly complex proposition and there is certainly no index at present that would allow sufficiently sensitive comparison between alternative treatment protocols. Clinical trials will focus more on “proximate” outcomes. These will mainly represent different aspects of anatomical form and function in the parts affected by the CFA, often reflecting the particular interests of individual provider groups. In essence, most measures will be an indication of the deficits that persist despite (or as a result of) treatment, such as shortcomings in appearance, speech, sight, hearing and dentofacial development. The general rules of reproducibility and validity apply, the latter being especially important when outcome is assessed before maturity. Longitudinal archives may be useful to determine the reliability of prediction for outcomes that are to be measured in the young (Shaw and Semb, 1996; Atack et al., 1997). Meaningful ways to document the satisfaction of patients and their families are essential, but present scales are rudimentary and may possess little validity. The development of techniques that have cross-cultural international validity has not begun and will be a significant challenge. In relation to cleft surgery, experience with a number of outcome measures and scales have been obtained regarding speech, dentofacial outcomes and patient satisfaction (e.g. Kuehn and Moller, 2000; Sell et al., 2001; Williams et al., 2001). Further work is certainly needed to refine these and build 26 Global strategies to reduce the health-care burden of craniofacial anomalies BOX C Systematic planning for surgical trials Whereas hypotheses for clinical trials in many disciplines will frequently be generated by laboratory-based studies or a consideration of previously reported cohort studies and clinical trials, this is unlikely to be the case for surgical trials in CFA surgery, at least for some time. Animal studies can shed some light on the general consequences of scars in the palatal mucoperiosteum, for example, but inferences for human maxillary growth are questionable (Kremenak, 1984; Friede, 1998; Leenstra et al., 1999). Furthermore, speech, a key outcome for cleft surgery is a uniquely human behaviour. The opportunity for most surgeons to gain meaningful experience of different techniques is severely constrained by the relative rarity of CFA subtypes, the need for lengthy follow-up, and the lack of robust measures of outcome.Together with the probable biases that apply to the existing CFA literature, research planning may be very idiosyncratic. In the absence of relevant animal studies and reliable clinical studies a process of informed negotiation would assist in defining promising alternatives in CFA surgery and in achieving the equipoise that must be established if clinicians are to enter ethically-grounded trials. By further negotiation, variations in current practices among potential partners could be harmonized/rationalized to create more manageable aggregations of trialists. One solution would be adoption of a focus group process supported by literature review specialists. Members of the focus groups would be selected on the basis of their knowledge and experience in the field, and their standing; the latter to encourage maximum credibility of the process and foster wide implementation of eventual trial findings.They would also be selected on their likely willingness and ability to enter and/or recruit surgical centres for the eventual trial. Collectively the focus groups should represent a good geographic and multidisciplinary spread. For different clinical topics such a process would define promising therapies, appropriate outcome measures, randomization schemes, and potential partners to develop cooperatives and funding applications. consensus upon international standards. Reliable rating of appearance is still problematical and, for speech, linguistic differences represent a significant international challenge. Outcomes should be patient-centred, i.e. measuring things that matter to ordinary people, rather than sophisticated surrogate measurements that may have little relevance to everyday life. Indeed, measurements of aesthetic and functional outcomes in isolation are not good predictors of emotional (psychological) adjustment and wellbeing (Robinson, 1997). There is a pressing need to identify the variables 27 WHO meetings on international collaborative research on craniofacial anomalies that contribute to the quality of life of affected individuals. Once identified, this knowledge should then be used to develop and refine methods of support and intervention, designed to optimize psychosocial as well as aesthetic and functional outcomes in CFA. BOX D Measuring treatment burden Since the consequences of CFA may be apparent through every phase of childhood and adolescence, there is seldom a time when the disciplines involved in care cannot recommend one or another intervention.The powerful desire of patients and parents to reach the point where the stigma of CFA will be completely eradicated makes it likely that they will accept most proposals and willingly comply with protocols of care recommended by all members of the team, no matter how demanding they may be. They have little choice. So far,“burden of care” has received little attention in CFA studies, yet the combined total of operations – other treatment episodes, and review appointments for the first 20 years of life, including all the disciplines that may be involved – can be enormous. Apart from pain and suffering and the disruption to family life, employment and school attendance, the dependent role in which this places the patient may have an adverse effect on the patient’s sense of self-determination or locus of control. A particular problem has arisen over the years with supplementary orthodontic interventions such as presurgical orthopaedics, primary dentition orthodontics and maxillary protraction. There is little evidence to suggest that the extra burden imposed on patients and the financial cost of these interventions is justified by any significant benefit (Severens et al., 1998; Long et al., 2001). Thus it is important in clinical trials to accurately record the total number of ancillary interventions and clinical visits in addition to surgical episodes. Measuring cost-benefit Economic pressures around the world have forced close examination of the true financial costs of treatment and, with reducing budgets, clinicians must either be involved in cost controls or have arbitrary choices imposed upon them. Surgical operations are invariably expensive treatment episodes and successful initial operations that minimize the need for multiple secondary revisions are highly desirable. Furthermore, successful initial repairs are likely to reduce the duration and complexity of subsequent ancillary procedures. Work has yet to begin in applying the techniques of health economics to the field of CFA. Health status and the utility of care and associated quality 28 Global strategies to reduce the health-care burden of craniofacial anomalies of life may be estimated using the techniques of time trade-off and conjoint analysis (Torrance, 1976; Ryan et al., 1998; Ryan, 1999). Economic prioritization models use decision analysis and simulation to assess the resource costs and patient benefits of current treatment patterns and the “cost-effectiveness gap” or potential gain from alternative surgical procedures for CFA. This would include reviews of existing literature, observational and audit databases to determine: the natural history of CFA; the incidence and prevalence of CFA; the possible indications and target populations for surgery; current treatment patterns and relevant comparators; and the costs and benefits of current treatment. Prospective registries occupy an intermediate position between non-randomized studies and randomized controlled trials Prospective registries – a preliminary approach for rare and/or novel interventions During the introductory phase of a new therapy it may be impossible to mount a randomized trial if the intervention is undergoing constant modification and the population it is applied to is heterogeneous and illdefined. Such is currently the case with many CFA interventions. A case in point in the last decade is distraction osteogenesis (gradual mechanical elongation of a bone) in its increasing application to the craniofacial skeleton. Pending the conduct of clinical trials, the establishment of prospective registries to enable critical appraisal of different kinds of CFA interventions will maximize collective experience and minimize the biases that inevitably occur with ad hoc reporting. Such registries would therefore play a similar role to Phase I trials of pharmaceutical interventions. One such registry has been set up for distraction osteogenesis in Europe as part of the EUROCRAN programme, with centres submitting duplicate records prior to – as well as after – treatment, as a step to minimizing follow-up, analysis and reporting bias (www.eurocran.net). As records of all cases would be filed with the registry prior to the start of treatment as well as after it, justification for non-follow-up would be required. And, as in well-conducted clinical trials, analysis bias could be overcome by employing blinded independent raters, while reporting bias could be overcome by the greater impartiality of the partnership and its predetermined conventions. Susceptibility bias and exclusion bias could not be minimized with the assurance derived from random allocation, but some checks of equivalence might be possible. Clinical proficiency, however, would inevitably remain as a major bias. Thus, prospective registries occupy an intermediate position between non-randomized studies and randomized controlled trials. 29 WHO meetings on international collaborative research on craniofacial anomalies The registry approach will maximize opportunities for preparatory work on outcome methodology: for early detection of extremely promising or unpromising clinical strategies, for defining answerable questions amenable to clinical trials, and for building the interpersonal trust and institutional partnerships that will be necessary to mount such trials. 3.1.4 Tissue engineering Tissue engineering offers two generic approaches to assist reconstruction Surgical advances of a more general, fundamental nature hold promise for improved CFA surgery in the foreseeable future. The discovery that, for example, wounds incurred during early gestation heal perfectly with no scars has led to intensive research of the cellular and molecular differences between scar-free healing and scar-forming healing (Whitby and Ferguson, 1991; Shah et al., 1992, 1996; Ferguson et al., 1996; Cornelissen et al., 2000a, 2000b, 1999a, 1999b). Thus the identification of high levels of TGFβ 3, with low levels of TGFβ 1 and 2, in scar-free wounds has led to the development of pharmaceutical interventions to reduce scarring in experimental skin wounds (e.g. www.renovo-ltd.com). Such interventions are currently undergoing trials in human volunteers and could offer considerable therapeutic benefits in surgery for cleft lip and palate and other CFA. A major problem in the surgical treatment of CFA is the deficiency of tissue available for surgical repair – bone, muscle, mucosa or specialized dental or eyelid tissues. Tissue engineering offers two generic approaches to assist reconstruction: either to grow cells outside the body, usually harvested from biopsy specimens, or to apply some form of scaffold to orientate the repair potential of the patient’s own cells in situ. Both approaches can be combined and it is now recognized that many of the cells participating in repair processes are stem cells, derived principally from bone marrow. Sophisticated scaffolds can be custom-made for the individual patient by defining the anatomical defect through three-dimensional reconstruction of CAT scan and MRI images and linkage to a prototyping or milling machine to manufacture a scaffold for the precise defect. Even the most delicate microsurgery is unable to accurately restore the muscle deficiencies of clefts of the lip and palate, but there is the prospect of encouraging muscle growth along a template of the body’s own proteins or a biodegradable polymer. Signalling by growth-factor release will enhance migration. Biomaterial science offers a potential solution for certain mechanical problems in CFA. Bone distraction techniques are effective in inducing bone formation and may be combined with osseointegration devices to allow longer-term movements of hard tissues. Detailed knowledge of 30 Global strategies to reduce the health-care burden of craniofacial anomalies internal stress analysis can be combined with cellular reactions to forcemechanotransduction to provide information to direct growth and tissue movement. The establishment of experienced clinical trial cooperatives will be essential to the safe, efficient and critical translation of these technologies into common practice. 3.1.5 Research on treatment BOX E Priorities for research on treatment There is an urgent need for the creation of collaborative groups in order to assemble a critical mass of expertise and to sufficiently access large samples of patients for adequately-powered clinical trials. Given the currently poor state of evidence for virtually all aspects of clinical management, there is an almost unlimited list of trials that could be initiated. However, the following were considered to be especially important: • trials of surgical methods for the repair of different orofacial cleft subtypes, not just unilateral clefts; • trials of surgical methods for the correction of velopharyngeal insufficiency; • trials of the use of prophylactic ventilation tubes (grommets) for middle-ear disease in patients with cleft palate; • trials of adjunctive procedures in cleft care, especially those that place an increased burden on the patient, family or medical services, such as presurgical orthopaedics, primary dentition orthodontics and maxillary protraction; • trials of methods for management of perioperature pain, swelling and infection; and nursing; • trials of methods to optimize feeding before and after surgery; • trials addressing the special circumstances of care in the developing world in respect of surgical, anaesthetic and nursing care; • trials of different modalities of speech therapy, orthondontic treatment and counselling. Equally urgent is the need to create collaborative groups, or improve the networking of existing groups, in order to develop and standardize outcome measures; there is an especially urgent need for work on psychological and quality of life measures, and economic outcomes. For rarer interventions, prospective registries should be established to hasten collaborative monitoring and critical appraisal, equivalent to Phase I trials. Relevant topics would be craniosynostosis surgery, ear reconstruction, distraction osteogenesis for hemifacial macrosomia and other skeletal variations, midface surgery in craniofacial dysostosis, and correction of hypertelorism. 31 WHO meetings on international collaborative research on craniofacial anomalies 3.2 Quality improvement Previous research demonstrates that similar interventions achieve widely different outcomes dependent upon the manner and circumstances in which care is provided. For example, secondary complications have been found to occur up to 10 times more frequently when the care of children with unilateral cleft lip and palate is performed inexpertly or delivered in an uncoordinated manner (Bearn et al., 2001). It is evident, too, that simple care can achieve equivalent or superior outcomes to complex care at less human and economic cost (Shaw et al., 1992b; Severens et al., 1998). The exploration of methods to define attainable standards of care for CFA and to promote quality-improvement protocols among the providers of care was considered to be an important priority. 3.2.1 Organization of services Delegates discussed the programme of quality-improvement activity conducted under the auspices of the European Commission between 1996-2000 (Shaw et al., 2001). This activity revealed great variability between countries in the provision of medical services for individuals with cleft lip and/or palate. While long-standing high-volume centres of expertise prevailed in Scandinavia, countries such as Italy, Germany, Switzerland and (until recently) the United Kingdom, provided cleft care via large numbers of local services with small case-loads. In other countries, such as Greece, Portugal and Spain, the concept of comprehensive specialist-team care was still undeveloped. The challenge of improving services in a pan-European manner was addressed in part by the consensual development of clinical and organizational guidelines. The difficulties observed in configuring services into specialized units with sufficient case-loads to foster proficiency of care and secure adequate resources for comprehensive care were by no means solely economic. Instead, the obstacles were frequently reported to be: personal egotism of individuals unwilling to discontinue the practice of treating a few children each year; competition between specialities for pre-eminence in the field e.g. plastic versus maxillofacial versus paediatric versus ear, nose and throat (ENT) surgery; local pride, with every hospital, town or region desiring its own small team; lack of clinical leadership; lack of responsiveness of the health authorities at local and national level. 32 Global strategies to reduce the health-care burden of craniofacial anomalies It was also noted that all the above problems had confronted the United Kingdom in the recent past and were not resolved until a national review was instigated by a government body (Sandy et al., 2001). The review included a national survey that revealed that Britain’s fragmented, decentralized services were achieving a low standard of clinical success. As a result the government instructed regions to provide care from a single regional centre, with a fully comprehensive specialist team – typically with two to three surgeons – each responsible for not less than 40-50 new personal cases requiring primary surgery per year. In this instance, government intervention was essential to the improvement of services when voluntary methods failed (Sandy et al., 2001). There is great variability between countries in the provision of medical services Elsewhere in Europe it was noted that the consensual guidelines on policies, practice guidelines and record-keeping had also been a powerful force in promoting reorganization of services for orofacial clefts, suggesting the influence of peer pressure at a national level. Thus within months of the publication of the European guidelines, more than half the countries in Europe had reconfigured services, formed new multidisciplinary collaborative associations, or increased funding for clinical services (Shaw et al., 2001). 3.2.2 International recommendations BOX F International recommendations on organization of cleft lip and palate services Delegates discussed the desirability of global recommendations on the principles that should govern clinical services for clefts of the lip and/or palate, and concerning basic clinical record collection. It was concluded that such guidelines would improve clinical research capability, and also encourage improved clinical care. There was special recognition of the economic constraints that would be faced by developing countries in complying with generic guidelines, but it was felt that these were still desirable to serve as a long-term goal. In particular, a set of guidelines recently developed through international consensus in Europe was reviewed. Delegates felt that these were appropriate as a basic requirement for wider international use and that the protocols recommended for clinical record collection were also acceptable as a minimum requirement. The recommendations of the WHO consensus conference are set out in Section 8. The rationale for recommending case-loads of 40 or more cases per operator is largely one of statistical imperatives: comparative clinical audit and research require adequate samples of cases with a similar prognosis. Clefts of the lip and palate present with great heterogeneity, and the only 33 WHO meetings on international collaborative research on craniofacial anomalies substantial category that is reasonably homogeneous is non-syndromic unilateral cleft of the lip and palate (UCLP). Even this group has considerable between-case variation, and reasonably large samples are required for statistical comparison. The Eurocleft Report (Shaw et al., 1992a) provided estimates of the sample sizes required to detect differences for a variety of outcomes. The Goslon Score, a rating of dental arch relationship (Mars et al., 1987) was found to require the lowest sample size for discerning differences among groups. One half point on the Goslon scale was the extent of the differences between the top- and middle-ranked centres and between the middle- and bottom-ranked centres in the Eurocleft study, equating to a 20% difference in osteotomy rate among such centres. At 5% probability and 80% power, detection of a 0.5 Goslon scale point difference in 10-year olds requires samples of the following size: 42 UCLP cases required in a 2-group comparison; 63 required in a 5-group comparison with 1 standard; and 77 required in a 6-group mutual comparison. Based on an occurrence of one non-syndromic complete unilateral cleft of the lip and palate, per six clefts of all types, Table 5 (below) shows the time it would take for surgeons, with a differing annual volume of cleft work, to generate varying samples. Table 5: Years required for the generation of samples of UCLP, related to case-load Surgeon volume 2-group comparison (n = 42) 42 8 4 Years to accrue sample for comparison 5-group 6-group mutual versus standard comparison (n = 63) (n = 77) 63 12 6 77 15 7.5 Source: Bearn et al., 2001 6 cases per year 30 cases per year 60 cases per year Even if follow-up is restricted to 5 rather than 10 years or more, it is clear that only operators treating 60 new cases per year would be able to audit their outcome within a decade. In the case of the United Kingdom, the figure of 40 cases per year (requiring approximately 12 years for an audit cycle) was the compromise reached. 34 Global strategies to reduce the health-care burden of craniofacial anomalies 3.2.3 Monitoring outcomes Participants agreed upon the desirability of establishing international standards, such as the development of rating methodology and samplesize estimates for comparison studies in the procedures of outcome evaluation, a process that also has a research dimension. Currently two general approaches were identified: Inter-centre comparisons: These might take the form of blinded nte -centr compariso isons: comparison of records of consecutive cases from different centres, a number of which have been reported (see Section 3.1.2). Alternatively, one set of records may be compiled to serve as a standard reference archive against which any team could compare its outcomes. A “good practice” archive of this kind might include durable records such as study casts, radiographs, speech tapes and so forth that would be representative of the ethnic population treated by well established teams with consistent protocols. Other teams could measure their own outcome records against these. In time a series of such archives for clefts and other CFA from different regions could become a webbased resource. The development of such an archive for Europe is included in the EUROCRAN programme (see Annex 2). In either case the recommended timetable for record collection would be helpful to maximize the opportunity for teams to successfully match their records to those from other centres (see Annex 5). istr R e g ist r ies: Under the auspices of the American Cleft PalateCraniofacial Association, a web-based “Craniofacial Outcomes Registry” (COR) was recently established, enabling North American teams to anonymously enter diagnostic and outcome data. Teams rate their own outcomes and can obtain an indication of their relative success compared with the Registry’s aggregated data (www.cfregistry.org). A national registry for the Cranofacial Anomalies Network in the United Kingdom has also been established and is developing protocols for standardized outcome data collection (www.perinatal.org.uk/crane). The Swedish Cleft Palate Association also has a web-based registry (Swedish National Quality Registry for Cleft Lip and Palate Treatment, http://natqa.uas.se/LKGreg/LKGreg.ihtml). It is intended that teams will display the actual records of consecutive cases, allowing peer review by each other. Participants in the meeting considered that joint, international work, in an effort to harmonize these differing approaches, was urgently required. Only operators treating 60 new cases per year could audit their outcome within a decade 35 WHO meetings on international collaborative research on craniofacial anomalies 3.3 Access and availability For the majority of CFA-affected individuals, there is a full life expectancy The meeting’s attention was drawn to the fact that, by the early 1960s, most industrialized countries had gained control of diseases caused by infection and/or malnutrition, and genetic disorders and birth defects had attained public health significance (Christianson, 2001). This situation is considered to occur when the infant mortality rate (IMR) falls below 40-50/1000 live births, at which juncture countries tend to recognize the need for medical genetic services. Approximately 40 years later, a significant proportion of the world’s developing nations has attained a similar situation: in 1997, 75 (53%) of the developing world’s countries, in which 60% of their population resided, had an IMR of less than 50 per 1000 live births. Only a minority of CFA are lethal and, for the majority of affected individuals, there is a full life expectancy. Appearance, function and social integration can, in nearly all cases, be improved by surgery and related multidisciplinary specialist medical care. The cost of treatment through infancy, childhood and beyond can be considerable however and, in the developing world, often unaffordable. For example, in 1994, the medical costs of one individual with cleft lip/palate in the United States was estimated at US$ 101 000 (Waitzman, 1994). In the United Kingdom, the estimated cost of 1 regional multidisciplinary cleft lip and palate service, receiving 140 new cases annually, is UK£ 6.4 million per year, excluding capital costs (National Health Service, United Kingdom, 2001). The social costs of unmet or partially-met medical needs are also enormous. Affected individuals are liable to suffer stigmatization, social exclusion and barriers to employment. When malnutrition and communicable diseases represent more pressing priorities, CFA care provided by nongovernmental organizations (NGOs), through charitable missions of medical staff or the external sponsorship of local providers, may be the only chance of treatment many individuals will have. Such efforts are known to be taking place on a remarkably large scale and in a wide variety of ways. Because of the distinctive features of these services it was considered that particular research questions need to be addressed in order to maximize the benefit of NGO endeavours in CFA. For example, in developing countries, patients often present for surgery at later ages than in developed countries, the services themselves may be of a rudimentary nature, and patients may be seen only once. Thus, a sound evidence base is needed to maximize effectiveness, safety and capacity. Again, quality-improvement strategies should be considered alongside this. 36 Global strategies to reduce the health-care burden of craniofacial anomalies 3.3.1 Main approaches Three main approaches to the provision of specialist care in the developing world were noted. The first was the establishment of efficiently run, high volume, indigenous centres of excellence, capable of serving large and widespread populations via a mixture of assisted travelling arrangements and outreach satellites. An example of such a centre that had achieved considerable success, both in providing service and conducting research, was presented (www.centrinho.usp.br). Secondly, some NGOs assist large numbers of individuals to receive surgery by providing financial support for indigenous clinical units to undertake operations that could not otherwise be afforded. Support for training indigenous specialists may also be provided (e.g. www.smiletrain.org). Thirdly, a large number of NGOs provide care by forming surgical missions where teams of surgeons and ancillary staff make visits to selected sites where there is a shortage of resources or experienced personnel (e.g. www.operationsmile.org; www.rotaplast.org). In several instances valuable research, especially of a genetic or epidemiological nature, has been conducted alongside these ventures (Lidral AC et al., 1997; Murray JC et al., 1997). Ethical issues are a prominent concern in this work and some programmes have been criticized on grounds of safety, surgical competence and absence of follow-up. Though not a research issue per se, it was felt that the present research programme taking place under WHO auspices should attempt to encourage agencies involved in the charitable provision of treatment in the developing world to develop and adhere to a common international code of practice. Such an effort might build upon the survey undertaken by an earlier international task force on volunteer cleft missions (Yeow et al., 1997). Agencies ... should be encouraged to develop and adhere to a common international code of practice 3.3.2 Further work Participants identified several areas deserving further work: a survey of the charitable organizations involved and the scale of their work; an appraisal of the cost-effectiveness and clinical effectiveness of the different models of aid; the promotion of dialogue between different NGOs to develop commonly-agreed codes of practice and adoption of the most appropriate forms of aid for local circumstances, with an emphasis on support that favours indigenous long-term solutions; 37 WHO meetings on international collaborative research on craniofacial anomalies the initiation of clinical trials concerning the specifics of surgery in a developing country setting: one-stage operations, optimal late primary surgery, anaesthesia protocols (e.g. local anaesthetic, inhalation sedation, antisepsis); the development of common core protocols for genetic, epidemiological and nutritional studies alongside surgery. 3.4 Regional perspectives A regional “good practice” reference archive for sub-Saharan Africa would be valuable The membership of the meeting was not intended to be fully representative of all nations. Several general observations, however, are possible, based upon the information presented. frica: Africa: In sub-Saharan Africa clinical resources for CFA are scarce as a consequence of prevailing economic problems and the greater challenge of communicable diseases, particularly AIDS. For example, in Namibia despite a high reported incidence, there are no cleft surgeons. As the wealthiest sub-Saharan country, South Africa has around 12 centres that undertake cleft surgery but these tend to work independently without common quality-improvement protocols. There has, as yet, been little formal study of CFA in the African population of sub-Saharan Africa and a regional “good practice” reference archive for this region would be valuable. There are a number of centres in the cities of Northern Africa but, as elsewhere in Africa, a survey has yet to be undertaken to identify potential sites with capability for collaborative research. ustr Ne Zealand: Austr alia and Ne w Zealand: There are well-developed services in many cities, though in some instances, the case-load is quite low, limiting the potential for collaborative research. However, the establishment of the Australian and New Zealand Craniofacial Association makes coordination possible and one centre has a programme of support and development for Indonesian and Malaysian cleft centres. C hina: In China there is reportedly a high level of unmet need for cleft and other CFA treatment. There is, however, a network of several large surgical centres that could form a potential research partnership. Treatment, however, is not free and follow-up is difficult. Speech therapists are especially scarce. Of those individuals receiving cleft surgery, only 30% are operated in the first year of life. Again this points to a need for surgical trials to define preferred operative techniques in more mature patients. A survey of clinical services and potential collaborating sites would be valuable, as would development of a quality-improvement strategy and “good practice” archive. 38 Global strategies to reduce the health-care burden of craniofacial anomalies uro Euro pe: European clinical services have recently been surveyed (Shaw et al., 2001). In the main, Europe’s problems arise from fragmentation of care over numerous small centres. The adoption of consensus recommendations, however, has begun to bring about restructuring, at least for cleft services. Several international research collaborations are under way (see Annex 1) and, under the EUROCRAN programme that was initiated in 2001, the European Commission is funding a series of multinational work packages that would be capable of wider networking (see Annex 2). sub ntine ubc inent: Indian subcontinent: As yet the subcontinent has not been surveyed regarding CFA or cleft services and research capability. However, an overview of India was presented and may be reasonably representative of adjoining countries. There are high levels of unmet needs and access is complicated as the majority of the population live in rural communities. There are several hundred surgeons trained in cleft surgery and several large university hospitals but, as yet, no quality-improvement protocols are in place. The subcontinent undoubtedly has numerous potential partners for clinical trials though resourcing follow-up studies will be a challenge. Latin mer Caribb arib Latin Amer ica and the Caribb ean: As yet no survey has been done on clinical services and research capability across the continent. Mexico was represented and has at least one large centre that has successfully completed clinical trials (Ysunza et al., 1998, 2001; Pamplona et al., 2001), and is recognized as a centre of excellence in the region. Brazil was also represented by the centre of excellence at Bauru. Elsewhere in Latin America there is undoubtedly a high level of unmet need. Sout-East Sout-East Asia: Singapore has already embarked upon a surgical trial in collaboration with a large centre of excellence in Taipei (www.nncf.org; www.cgmh.org.tw) and together they have a high research capability. In Indonesia there are high levels of unmet need but around six cleft teams are established and would be potential sites for research collaboration. Already both Indonesia and Malaysia are engaged in epidemiological, nutritional and genetic research with agencies in Australia, Europe, Singapore and elsewhere. There are reportedly high local incidences of CFA, such as frontal encephalocele, that may be fruitful targets for multidisciplinary research. Like Europe, Japan may have a fragmentation of services in small centres; however, the Japanese Cleft Palate Association has begun discussions on inter-centre studies and clinical trials. In Korea, several high-volume centres are potential sites for collaborative research and the Korean Cleft Palate Association has begun discussion on inter-centre studies. The Indian subcontinent has numerous potential partners for clinical trials 39 WHO meetings on international collaborative research on craniofacial anomalies iddle M iddle East: A high level of unmet need has been reported with few established CFA centres. A number of university hospitals in the region would be potential partners in research. mer Nor th Amer ica: North America also suffers from a fragmentation of cleft and craniofacial services, and representatives from there spoke of the difficulties of obtaining sufficient subjects for clinical trials because of the decentralized nature of services. The recent emergence of health management organizations was seen as a particular force for the fragmentation of services and dissipation of established cleft teams. None the less, the Childhood Cancer Study Group has achieved a high level of coverage in the United States, as a result of which a high proportion of affected children are enrolled in trials (Ross et al., 1996; Shocat et al., 2001). The American Cleft Palate-Craniofacial Association has promoted adequate team care and has published several sets of guidelines, as well as initiating the Craniofacial Outcomes Registry. 40